The Baltimore Longitudinal Study of Aging (BLSA) was established in 1958 and is one the oldest prospective studies of aging in the USA and the world. The mission of the BLSA is to learn what happens to people as they get old and how to distinguish changes due to aging from those due to disease or other causes. The Early Markers of Alzheimer's Disease program continues to perform cognitive assessments and establish research diagnoses of Mild Cognitive Impairment and Alzheimer's Disease for BLSA participants. This information is used in multiple collaborative research projects conducted by intramural and extramural investigators, including our studies of brain aging and neuroimaging biomarkers of cognitive decline and AD. Over the last year, we have continued cognitive assessments of BLSA participants, as well as diagnostic case conferences, to establish research diagnoses of cognitive impairment and to define age-associated cognitive changes. We have continued to investigate possible modifiers of cognitive aging, risk for dementia, and the presence of Alzheimer's pathology at autopsy. In a recent publication (Williams et al., Alzheimers & Dementia, in press), we performed sex-stratified analyses to determine how Apolipoprotein E (APOE) e4 genotype and age interact to influence longitudinal decline in different cognitive domains. Longitudinal cognitive trajectories were examined in cognitively normal Caucasian older adults (631 men, 561 women, baseline age range: 50-93, 6733 assessments). In men, older baseline age was associated with greater effects of APOE-e4 on longitudinal decline in memory and executive function, detectible from baseline age of 64 and 68, respectively. In women, older baseline age was associated with greater APOE-e4 effects on longitudinal decline in attention, detectible at baseline age of 66. No significant APOE-e4 effects were found for language, visual-spatial, ability or processing speed. Our results highlight the importance of considering sex and age when assessing APOE-e4 associated vulnerability to cognitive decline. In a second paper based on a collaborative study with Dr. Ellen Grober (JINS 2019), we examined trajectories of declines in learning and retention during the predementia phase of Alzheimer's disease (AD) using the picture version of the Free and Cued Selective Reminding Test with Immediate Recall (pFCSRT+IR). Learning was defined by the sum of free recall over three test trials, and retention was defined by delayed free recall (DFR). Repeated assessments of 217 incident AD cases from the BLSA Early Markers study were aligned based on the time of AD diagnosis. The predementia phase of learning and retention decline was assessed using change point models in which cognitive trajectories are described by a series of linear components with knots delineating times of accelerating decline. Results demonstrated two change points for both learning and DFR: the first at 6.58 (95% confidence intervals (CI): 6.56, 6.60) to 7.29 (95% CI: 6.13, 8.46) years before diagnosis followed by gradual decline over the next 4 years, and a second acceleration of decline 1.89 (0.56, 3.24) to 2.93 (95% CI: 1.56, 4.30) years before diagnosis. The change points for DFR were not significantly earlier in the predementia phase than the change points for learning. These results showed that both learning and DFR had similar profiles of decline in the years prior to a the clinical diagnosis of AD. We also continue our collaboration with the Johns Hopkins Alzheimers Disease Research Center. In a recent study, BLSA autopsy material from the brain tissue bank was used in conjunction with antemortem cognitive measures to characterize individuals who meet the diagnosis of primary age-related tauopathy (PART). PART is a recently described entity that can cause cognitive impairment in the absence of Alzheimer's disease (AD). We compared neuropathological features, tau haplotypes, APOE genotypes, and cognitive profiles in age-matched subjects (N=183, age 85 years and older) with PART and AD pathology. Sixteen of 34 (47%) cases with PART versus 90 of 111 (81%) cases with AD pathology met consensus diagnosis of cognitive impairment. Compared with AD, PART subjects showed significantly slower rates of decline on measures of memory, language, and visuospatial performance. They also showed lower APOE e4 allele frequency (4.1% vs. 17.6%, P = .0046). Our observations suggest that PART is separate from AD and its distinction will be important for the clinical management of patients with cognitive impairment and for public health care planning. We continue to collaborate extensively with both intramural and extramural investigators, including ongoing studies of motor function, sleep disturbance, hearing loss, and vision changes. Over the last year, we also have continued an active collaboration with a 5-study consortium of longitudinal studies focused on preclinical AD. Participants at the study sites are included in the consortium if they are cognitively normal at baseline, have either PET-PiB or cerebral spinal fluid measures of amyloid-beta and serial cognitive assessments. The collaboration and larger sample sizes provided by the 5 studies and the University of Pennsylvania image analysis core will allow tests of more complex interactions influencing risk and protective factors for Alzheimer's disease during the preclinical asymptomatic stage.